Research on improving democracy for poor citizens

In democracies around the world, poor people and members of minority groups have won the right to vote – yet their elected officials often work against their interests. Why does this happen, and can greater transparency give the poor and marginalized a stronger voice in democracy?


In Delhi, the anticipation of publicly disclosing politician performance prior to election led to an increase in pro-poor spending in high-slum areas, and to alternative electoral opportunities for high-performing incumbents who were rendered ineligible by quotas; there was no increase in spending among politicians who were located in low-slum areas nor among politicians whose performance was not publicly disclosed.


Since 2008, researchers from the Inclusion Economics team – with collaborators from Columbia University, Harvard University, and MIT – have worked to explore the effects of giving people better information about the performance of their elected politicians – to see whether increased transparency can change election outcomes, patterns of public spending, and politicians’ behavior. In one intervention, researchers distributed pamphlets about the importance of voting and newspapers with report cards evaluating the performance of the locally elected representatives in a Delhi slum area. Analysis showed that households receiving these materials used the information to compare their electoral options and change voting patterns. Local politicians also changed their behavior when told that their performance would be disclosed – but not when they received this information privately. For a relatively small transparency intervention, such measurable effects on policies and elections are notable. This research project is complete – having generated a published paper, a submitted working paper, and a third working paper – and its findings could be used to improve elections and strengthen civic engagement across India and beyond.

Related Publications

Public Information is an Incentive for Politicians: Experimental Evidence from Delhi Elections

In 2010, authors Banerjee, Enevoldsen, Pande, and Walton informed a random set of Delhi councilors that a newspaper would report on their performance. The research team found that councilors who were informed about performance reporting increased pro-poor spending in high-slum wards. The findings suggest that the public nature of report cards, not access to information on public services per se, incentivized councilors; and that, in low-information situations, credible public disclosures of politician achievements matters to both parties and voters.

Are Poor Voters Indifferent to Whether Elected Leaders Are Criminal or Corrupt? A Vignette Experiment in Rural India

Using 2010 data from the Indian state of Uttar Pradesh, authors Banerjee, Green, McManus, and Pande explored whether voters frequently elect and reelect criminal or venal candidates due to the patronage they provide. The study found that voters presented with information on the attributes of competing legislative candidates become much less likely to express a preference for candidates who are alleged to be criminal or corrupt.

Do Informed Voters Make Better Choices? Experimental Evidence from Urban India

In the run-up to the 2008 elections in Delhi, authors Banerjee, Kumar, Pande and Su implemented a field experiment providing slum dwellers with information on candidate qualifications and legislator performance. Access to report cards increased voter turnout, and this effect was found to be larger when incumbent performance was worse.